W3C home > Mailing lists > Public > w3c-sgml-wg@w3.org > November 1996

Re: XML, HTML, SGML, life, the universe, and everything

From: Steven J. DeRose <sjd@ebt.com>
Date: Mon, 11 Nov 1996 14:58:48 -0500
Message-Id: <>
To: W3C-SGML-WG@w3.org
At 10:36 AM 11/11/96 GMT, Charles F. Goldfarb wrote:
>Conforming SGML means that any conforming system will produce the same ESIS. At
>present, an XML document is not conforming SGML. 

Well, not exactly ESIS, since ESIS isn't part of 8879. 

Given that 8879 doesn't say what the result of parsing is, how do you tell
whether a conforming system has done "the same thing" as another one has,
*purely* given requirements of 8879? In 8879, a parser can be a black box
that returns TRUE or FALSE (and maybe some error messages). For example, is
there a normative definition for what it actually means to "ignore" an RE?
If we take "ignore" at its English face value, then an SGML system which
produces a byte-identical copy of its input is non-conforming -- it clearly
doesn't ignore the REs that 8879 requires be ignored, so it violates 8879.

What I'm saying is, if XML output falls so close to SGML's that nobody cares
about the difference in practice, did we make a noise?

Received on Monday, 11 November 1996 14:59:10 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:25:20 UTC